-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path3b_circDE.R
212 lines (174 loc) · 7.97 KB
/
3b_circDE.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
library(AnnotationDbi)
library(org.Hs.eg.db)
library(DESeq2)
library(tximport)
library(GenomicFeatures)
library(fst)
library(DOSE)
library(clusterProfiler)
library(tidyverse)
# import common functions
source("combinedLinCirc/PD-RNA/functions.R")
# IMPORT METADATA ---------------------------------------------------------
meta.ppmi <- read_rds("circRNA/data/ppmi_metadata.rds") %>%
mutate(
pct_intronic_bases = scale(pct_intronic_bases),
)
meta.icicle <- read_rds("circRNA/data/icicle_metadata.rds") %>%
mutate(
pct_intronic_bases = scale(pct_intronic_bases),
median_cv_coverage = scale(median_cv_coverage)
)
# IMPORT UNFILTERED AND DE DDS --------------------------------------------
bsj_dds.ppmi <- read_rds('circRNA/data/ppmi_bsj_DDS_preFilter.rds')
bsj_dds.icicle <- read_rds('circRNA/data/icicle_bsj_DDS_preFilter.rds')
fsj_dds.ppmi <- read_rds('circRNA/data/ppmi_fsj_DDS_preFilter.rds')
fsj_dds.icicle <- read_rds('circRNA/data/icicle_fsj_DDS_preFilter.rds')
# OVERWRITE DESIGN --------------------------------------------------------
# OG DESIGN JUST HAS CONDITION FOR QC AND SETUP
# specify design formula
design.ppmi <- ~ sex + batch + age_bin + pct_intronic_bases + condition
design.icicle <- ~ sex + batch + age_bin + pct_intronic_bases + median_cv_coverage + condition
bsj_dds.ppmi@design <- design.ppmi
bsj_dds.icicle@design <- design.ppmi
fsj_dds.ppmi@design <- design.ppmi
fsj_dds.icicle@design <- design.icicle
# PREFILTERING ------------------------------------------------------------
# filter BSJs based on a filtering strategy
filterDDS <- function(dds, metadata) {
# apply raw count filtering
keep <- rowSums(counts(dds, normalized = FALSE) > 10) > min(table(metadata$condition))
# how many genes removed
genesRemoved <- function(keep, metadata) {
total <- nrow(data.frame(keep))
ntrue <- data.frame(logical = keep) %>%
filter(logical == TRUE) %>%
nrow()
nfalse <- data.frame(logical = keep) %>%
filter(logical == FALSE) %>%
nrow()
print(paste0("Total number of genes tested: ", total))
print(paste0("Number of genes kept: ", ntrue, " (", round(ntrue / total * 100), "%)"))
print(paste0("Number of genes removed: ", nfalse, " (", round(nfalse / total * 100), "%)"))
}
genesRemoved(keep, meta_ppmi)
dds <- dds[keep, ]
return(dds)
}
bsj_dds.ppmi <- filterDDS(dds = bsj_dds.ppmi, metadata = meta.ppmi)
bsj_dds.icicle <- filterDDS(dds = bsj_dds.icicle, metadata = meta.icicle)
# now filter the FSJ counts for the same junctions
fsj_dds.ppmi <- fsj_dds.ppmi[rownames(fsj_dds.ppmi) %in% rownames(bsj_dds.ppmi), ]
fsj_dds.icicle <- fsj_dds.icicle[rownames(fsj_dds.icicle) %in% rownames(bsj_dds.icicle), ]
# DIFFERENTIAL EXPRESSION -------------------------------------------------
getDDS <- function(dds, runOrImport, cohort, type) {
if (runOrImport == "run") {
cat("Running DESeq2 (might take a while)\n")
dds <- DESeq(dds)
write_rds(dds, paste0("circRNA/data/", cohort, "_dds_", type, ".rds"))
return(dds)
} else if (runOrImport == "import") {
cat("Importing previously run DESeq2 analysis DeseqDataSet\n")
dds <- read_rds(paste0("circRNA/data/", cohort, "_dds_", type, ".rds"))
return(dds)
} else {
stop("runOrLoad must be either run or import")
}
}
# DE - RUN OR IMPORT ------------------------------------------------------
# BSJ
bsj_dds.ppmi <- getDDS(dds = bsj_dds.ppmi, runOrImport = "run", cohort = "ppmi", type = "bsj")
bsj_dds.icicle <- getDDS(dds = bsj_dds.icicle, runOrImport = "run", cohort = "icicle", type = "bsj")
fsj_dds.ppmi <- getDDS(dds = fsj_dds.ppmi, runOrImport = "run", cohort = "ppmi", type = "fsj")
fsj_dds.icicle <- getDDS(dds = fsj_dds.icicle, runOrImport = "run", cohort = "icicle", type = "fsj")
# RESULTS -----------------------------------------------------------------
# import junc file
junc_info <- read_fst("circRNA/data/bound_juncInfo.fst")
# import junc info file to get annotations
coord_anno <- junc_info %>%
select(coord_id, gene_id, gene_name) %>%
distinct()
# set contrast level
contrast <- c("condition", "PD", "Control")
getDeseqResults <- function(dds, rowname) {
results <- results(dds, contrast = contrast, independentFiltering = FALSE, alpha = 0.05)
print(summary(results))
results <- results %>%
as.data.frame() %>%
rownames_to_column(rowname)
return(results)
}
# BSJ
bsj_results.ppmi <- getDeseqResults(dds = bsj_dds.ppmi, rowname = "coord_id") %>%
left_join(coord_anno, by = "coord_id")
write_csv(bsj_results.ppmi, "circRNA/output/ppmi_BSJresults.csv")
bsj_results.icicle <- getDeseqResults(dds = bsj_dds.icicle, rowname = "coord_id") %>%
left_join(coord_anno, by = "coord_id")
write_csv(bsj_results.icicle, "circRNA/output/icicle_BSJresults.csv")
fsj_results.ppmi <- getDeseqResults(dds = fsj_dds.ppmi, rowname = "coord_id") %>%
left_join(coord_anno, by = "coord_id")
write_csv(fsj_results.ppmi, "circRNA/output/ppmi_FSJresults.csv")
fsj_results.icicle <- getDeseqResults(dds = fsj_dds.icicle, rowname = "coord_id") %>%
left_join(coord_anno, by = "coord_id")
write_csv(fsj_results.icicle, "circRNA/output/icicle_FSJresults.csv")
exportFilteredCounts <- function(dds, cohort, type) {
# raw counts
counts(dds, normalized = FALSE) %>%
as.data.frame() %>%
rownames_to_column("coord_id") %>%
write_csv(paste0("circRNA/data/", cohort, "_deseq", type, "FilteredRawCounts.csv"))
cat(paste0(cohort, ": exported filtered raw counts"), "\n")
# normalised counts
counts(dds, normalized = TRUE) %>%
as.data.frame() %>%
rownames_to_column("coord_id") %>%
write_csv(paste0("circRNA/data/", cohort, "_deseq", type, "FilteredNormalisedCounts.csv"))
cat(paste0(cohort, ": exported filtered normalised counts"), "\n")
# VST counts
assay(varianceStabilizingTransformation(dds, blind = FALSE)) %>%
as.data.frame() %>%
rownames_to_column("coord_id") %>%
write_csv(paste0("circRNA/data/", cohort, "_vst", type, "Counts.csv"))
cat(paste0(cohort, ": exported VST counts"), "\n")
}
exportFilteredCounts(bsj_dds.ppmi, "ppmi", "BSJ")
exportFilteredCounts(bsj_dds.icicle, "icicle", "BSJ")
exportFilteredCounts(fsj_dds.ppmi, "ppmi", "FSJ")
exportFilteredCounts(fsj_dds.icicle, "icicle", "FSJ")
# # TEST (don't run) ----------------------------------------------------------------------------
# # BSJ limited covariates as test
# testDDS <- function(dds) {
# test_dds <- dds
# test_dds@design <- design <- ~ sex + batch + age_bin + condition
# test_dds <- DESeq(test_dds)
# test_results <- getDeseqResults(test_dds, "coord_id")
# return(test_results)
# }
# test_dds.ppmi <- testDDS(bsj_dds.ppmi) %>% left_join(coord_anno, by = "coord_id")
# write_csv(test_dds.ppmi, 'circRNA/output/ppmi_BSJ_results_limitedModel.csv')
# test_dds.icicle <- testDDS(bsj_dds.icicle) %>% left_join(coord_anno, by = "coord_id")
# write_csv(test_dds.icicle, 'circRNA/output/icicle_BSJ_results_limitedModel.csv')
# FUNCTIONAL ANALYSES -----------------------------------------------------
# significant BSJs
sig_bsj_enrich.ppmi <- goEnrich(
geneSet = bsj_results.ppmi %>% filter(
padj < 0.05,
log2FoldChange > 0.1 | log2FoldChange < -0.1
) %>% pull(gene_id),
background = bsj_results.ppmi %>% pull(gene_id),
ontology = "ALL"
)
saveRDS(sig_bsj_enrich.ppmi, "circRNA/output/ppmi_sigBSJ_enrich.rds")
# ABUNDANT BSJs
abundant_bsj_enrich.ppmi <- goEnrich(
geneSet = bsj_results.ppmi %>% pull(gene_id),
background = junc_info %>% filter(study == "PPMI") %>% pull(gene_id) %>% unique(),
ontology = "ALL"
)
saveRDS(abundant_bsj_enrich.ppmi, "circRNA/output/ppmi_abundantBSJ_enrich.rds")
abundant_bsj_enrich.icicle <- goEnrich(
geneSet = bsj_results.icicle %>% pull(gene_id),
background = junc_info %>% filter(study == "ICICLE-PD") %>% pull(gene_id) %>% unique(),
ontology = "ALL"
)
saveRDS(abundant_bsj_enrich.icicle, "circRNA/output/icicle_abundantBSJ_enrich.rds")